Measurements

In the KNHANES, after asking the participants if they have had a job during the last 1 year, the working schedule of the participants who ever had worked during the last 1 year was assessed. The major working schedules were classified as follows: mainly daytime work (6 am–6 pm), evening work (2 pm–midnight), night work (9 pm– 8 am the next day), day-night regular shift work, 24-hour rotating work, split shift, irregular shift work, or others. In this study, participants were considered as having day work if they had worked between 6 am and midnight in combination with daytime (6 am–6 pm) and evening work (2 pm–midnight), and night-shift if they had night work (9 pm–8 am next day) or day-night regular shift work. However, in the analysis, we divided night-shift work into two categories: night work and day-night regular shift work. Participants who reported other working schedules were excluded from the analysis due to the small number of participants. The baseline characteristics and lifestyles were recorded through interviewer-guided questionnaires, including age, gender, education, alcohol consumption, smoking status, average number of hours of sleep, stress, and physical activity.

The AUDs were assessed using the Alcohol Use Disorders Identification Test (AUDIT), with a scale from 0 to 40 [16]. Briefly, this test includes 10 questions related to hazardous alcohol use (frequency of drinking, typical volume, and heavy drinking), symptoms of alcohol dependence (ability to stop drinking, ability to control normal activities, and morning drinking), and harmful alcohol use (remorse after drinking, blackouts, alcohol-related injuries, and other concerns about drinking) as a simple method of screening for hazardous and harmful alcohol consumption in primary care setting [17]. We used the AUDIT cutoff value of ≥8 points to define AUDs. This was proven to give an appropriate level of sensitivity and specificity in unselected populations [16]. Additionally, based on the WHO guidelines, AUDIT scores were categorized into four zones with their corresponding interventions: zone I, 0–7 points: alcohol education; zone II, 8–15 points: simple advice; zone III, 16–19 points: counseling and continued monitoring; and zone IV, 20–40 points: referral to specialists for diagnostic evaluation and treatment [18].

HRQL was measured by EuroQol-5D (EQ-5D), which was developed by the EuroQoL Group. EQ-5D was descriptively quantified by five dimensions, namely mobility, self-care, usual activities, pain/discomfort, and anxiety/depression. Each of the dimension was described as follows: 1  =  “no problem,” 2  =  “some problems,” and 3  =  “severe problems.” All of these response levels were converted into an EQ-5D index using the weight scoring system of the five dimensions ranging from 0 (worst) to 1 (best) [19]. The five dimensions were used as dichotomous variables by merging levels 2 and 3 into “some or extreme problem.”